Yufei Huang
University of Pittsburgh, PA·Computer Science
Cells in a tissue are not scattered at random. Their positions and the genes they switch on together shape how an organ works and how disease shows up.
- Fields
Computer Science
Genetics
Biomedical Engineering
- Topics
spatial transcriptomics
single-cell spatially resolved transcriptomics
scSRT
- Grants
- 2 known grants, through Mar 2027
What they work on
- Develop a contrastive learning-based geometric graph attention model that learns gene expression patterns aware of tissue shape from single-cell spatial transcriptomics data, so that the patterns match the tissue structure seen in histology.
- Build a semi-supervised method that classifies points on a geometric graph to outline diseased tissue regions from spatial gene expression using only a few labels, and test it against annotated histology such as COVID-19 lung tissue.
- Create tools that detect and measure single-base m6A RNA methylation sites from MeRIP-seq and nanopore sequencing data, trained on more than 1,700 human and mouse samples.
- Predict when the m6A mark on RNA triggers RNA decay or changes splicing, and catalog these predictions for cancer and normal cells and tissues in m6A-KB, a searchable public knowledgebase.
More about the research
New technologies can now measure gene activity in single cells while keeping their place in the tissue, and in cancer, a chemical tag on RNA can change how genes behave. Both produce huge, complex datasets that need new computational tools. Across two projects, the lab builds machine learning methods and data resources to make sense of these measurements.
Why it matters
These tools could help researchers and pathologists automatically find diseased regions in tissue samples. They could also help uncover the cell and gene signatures behind newly emerging or poorly understood diseases, such as the lung damage seen in COVID-19. For cancer researchers, the m6A-Suite tools and the m6A-KB knowledgebase are meant to show how RNA methylation drives tumor growth, which could inform future m6A-based cancer therapies.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
m6A-suite: an informatics pipeline and resource for elucidating roles of m6A epitranscriptome in cancer
NIH · Apr 2023 – Mar 2027
Novel geometric deep learning models for tissue structure-aware spatial expression representations from spatially resolved single-cell transcriptomics data
NIH · Sep 2024 – Aug 2026
Closest labs
Philip Yu
University of Illinois Chicago, IL·Computer Science
Integrate open-source anomaly detection tools for tabular, graph and time-series data into a single OpenAD ecosystem with a standardized development environment for building and evaluating models.
Lingwei Chen
Rochester Institute of Technology, NY·Computer Science
Repurpose pretrained transformer models for cyberattack detection through cross-modal adversarial reprogramming, which adds patch-level changes to inputs so few new parameters are needed when labeled data are scarce.
Sunil Sudarshan
University of Alabama at Birmingham, AL·Biochemistry
Determine the molecular mechanism by which the oncometabolite L-2-hydroxyglutarate promotes the mRNA mark N6-methyladenosine (m6A) in renal cell carcinoma, and establish what this RNA hypermethylation does to gene expression.
The school
University of Pittsburgh: 83 labs on PI Pages, $1.5B a year on research, 403 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.